On March 8, Jay Butler, director of Realty Studies at Arizona State University’s Polytechnic campus, released his home sales report for February. Butler refrained from making any projections for the coming year. In an article by Betty Beard in the Arizona Republic Butler called February: [Hat tip L!]
A "pitiful month" for forecasting events because it, along with January, is usually one of the slowest months for sales of existing homes.
In his report he elaborated:
February, like January, tends to be a poor indicator of the coming year. In addition, it is a short month, so it is typically a low month for sales activity, frequently being the lowest month of the year. For February, 4,280 homes were recorded sold, in contrast to 4,520 for January, 5,460 for a year ago and 7,935 transactions in 2005. This was the lowest February since 4,090 homes were recorded sold in 2003, which was the lowest month for 2003. So far in 2007 a total of 8,800 homes have been recorded sold in contrast to 10,715 in 2006 and 17,290 in 2005 for the same time period.
I disagree with Butler’s assessment of February as a poor sales indicator. Here’s the sales graph from 2002-2007: [Sales numbers 2002-2006 from ARMLS, 2007 from Butler***]
Also note that with the exception of 2006, which cooled rapidly as the bubble began to fizzle, the years remain fairly consistent in their trends, with little crossover. It seems unlikely, based on the above, that 2007 sales will exceed 2002 levels. It is possible that souring buyer sentiment and tightening lending standards may in fact drive sales this year below 2002 levels. With inventory nearly back to record highs, this could magnify the current imbalance in the market, making things more difficult for sellers and increasing the rate of foreclosures. This will also continue to put downward pressure on home prices.
As for median prices Butler reports:
Much like the ever-increasing sales activity of the last few years, the rapid improvement in prices has disappeared. The median home price has been very stable at $260,000, which is the same as January, but down from last year’s $265,000. [-1.9% YOY – not adjusted for inflation]
In the Republic he also stated:
Compared with February 2006, median prices last month rose in Ahwatukee Foothills and Chandler and fell in Gilbert, Tempe and Mesa, according to realty studies’ monthly numbers.
It’s just one of those markets that has no trend.
I also disagree with Butler on the market trend. While some communities are still reporting an increased median price from last year, the general trend in the Valley is down:
"Bubble debunkers" sometimes claim that the bubble does not exist as it failed to "pop" as predicted by "bubble-mongers." Those who study speculative bubble markets though, generally do not predict a "pop," but a gradual rate of decline, particularly in real estate, where prices tend to be "sticky." This is in fact the trend we are seeing.
My own community of Gilbert fared much worse than the Valley average- it dropped from $341,000 to $307,500 for a 10% price decline.
Butler does venture a prediction for sales next month:
Listings tend to rise in January and February, so March numbers might show more sales. "Right now it looks reasonably good," he said.
The sales graph shows that generally the February-March sales increase is the greatest of the year. I would up Butler’s "reasonably good" to "virtually certain." Let’s hope so, as a decline would not be in keeping with market norms, and would not bode well for Valley home sales this year.
*** You will note that Butler’s description of previous Februarys does not match the graph. I work with three sets of data when looking at the Phoenix resale market. One would be the reports issued by ARMLS, another would be the reports issued by Jay Butler, and the other would be the numbers pulled directly from the MLS by really thoughtful and much appreciated agents! [I made a request to Jay Butler to see if he would explain the discrepancy, but received no response.]
While there generally is not a great deal of variation between the three, there is enough that I try to keep an "apples to apples" comparison and not compare one series to another. When that is not possible, I give an explanation of my sources, such as today.
The Butler data that is publicly available is spotty for 2003, and not available for prior years. I used the ARMLS data in order to show multiple years of sales. Only Jan and Feb 2007 are Butler’s numbers. I could not use ARMLS numbers for 2007, as their official number for February has not been released yet.